From bcpierce at cis.upenn.edu Mon Apr 4 12:51:36 2005 From: bcpierce at cis.upenn.edu (Benjamin Pierce) Date: Mon Apr 4 06:51:42 2005 Subject: [unison-hackers] Hosting issues. In-Reply-To: Your message of Sun, 03 Apr 2005 16:33:33 +0100. <033801c53862$7ed02d90$6508a8c0@chimaera> Message-ID: <3858.1112611896@wlan-025.wireless.net.ed.ac.uk> Hi Max, We can certainly set you up with some way of getting the files to Penn. I'm wondering if the best way to do this is to give you a login account on some local machine where you can upload files, or whether it could actually be done via subversion, by adding you to the list of developers (it's probably time to do that anyway!) and putting the binaries into the svn repository. I'm not (yet :-) an svn expert -- can someone comment on whether it is going to choke on large files like statically linked executables? - Benjamin P.S. Remember to post messages to unison-hackers@lists.seas.upenn.edu -- the old unison-hackers@yahoogroups.com lists is still there (I have not deleted it yet), but is now defunct. > > > > > > So far, I've been hosting my Windows compiled binaries of Unison on a
> machine at my university.
>
> In a few months, I will graduate, and no longer have this facility available
> to me.
>
> I'm less than enthusiastic about trying to find a free webspace service, so
> I'm hoping that I can be given access to a contrib directory on the official
> unison ftp site.
>
> Will this be possible?
>
> Max.
>
> > > > > > > >
>
> Yahoo! Groups Links
> > >
> > > > > > > From gav at thataddress.com Mon Apr 4 09:51:28 2005 From: gav at thataddress.com (George Vilches) Date: Mon Apr 4 08:51:51 2005 Subject: [unison-hackers] Hosting issues. In-Reply-To: <3858.1112611896@wlan-025.wireless.net.ed.ac.uk> References: <3858.1112611896@wlan-025.wireless.net.ed.ac.uk> Message-ID: <42513850.2070203@thataddress.com> Benjamin Pierce wrote: > > I'm not (yet :-) an svn expert -- can someone comment on whether it is going > to choke on large files like statically linked executables? I've run a few rather large repositories, and binary files of larger sizes seem to hold up just fine. The Subversion documentation makes a lot of noise about how good it is for binary files, so I'd say go for it. George From hvrietsc at yahoo.com Mon Apr 4 10:22:32 2005 From: hvrietsc at yahoo.com (HvR) Date: Mon Apr 4 12:22:43 2005 Subject: [Unison-hackers] clobberin gin star configurations Message-ID: <20050404162233.69553.qmail@web53705.mail.yahoo.com> About a year or two ago I set up a star configuration: many "clients" syncing to one central "server". At that time unison would create a .# temp file such that if two unisons from different systems would be syncing the same file at the same time they would each create the same .# temp file and thus clobber each other. At that time a special unstable version existed that would create a .# temp file and thus the clobbering would not happen. The current stable version does not seem to do this. Is this no longer needed? Or do we still need to fix this. __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From bcpierce at cis.upenn.edu Mon Apr 4 18:55:17 2005 From: bcpierce at cis.upenn.edu (Benjamin Pierce) Date: Mon Apr 4 13:28:28 2005 Subject: [unison-hackers] Hosting issues. In-Reply-To: Your message of Mon, 04 Apr 2005 14:30:12 +0100. <0bd101c5391a$6e7b6280$6508a8c0@chimaera> Message-ID: <4239.1112633717@benjamin-pierces-computer.local> > > P.S.  Remember to post messages to unison-hackers@lists.seas.upenn.edu --
> > the old unison-hackers@yahoogroups.com lists is still there (I have not
> > deleted it yet), but is now defunct.
>
> Was this announced? I wasn't aware of it. Plus, when I post to the upenn.edu
> address, I get a bounce:
>
> |You are not allowed to post to this mailing list, and your message has
> |been automatically rejected.  If you think that your messages are
> |being rejected in error, contact the mailing list owner at
> |unison-hackers-owner@lists.seas.upenn.edu.
Yes, it was announced a few weeks ago. I (thought I) copied the old membership list (did you get two copies of this?), but in any case you should be able to sign up for the list by visiting the home page http://lists.seas.upenn.edu/mailman/listinfo/unison-hackers -- Benjamin From bcpierce at cis.upenn.edu Mon Apr 4 19:35:31 2005 From: bcpierce at cis.upenn.edu (Benjamin Pierce) Date: Mon Apr 4 13:35:52 2005 Subject: [Unison-hackers] clobberin gin star configurations In-Reply-To: Your message of Mon, 04 Apr 2005 09:22:32 -0700. <20050404162233.69553.qmail@web53705.mail.yahoo.com> Message-ID: <4410.1112636131@wlan-068.wireless.net.ed.ac.uk> > About a year or two ago I set up a star configuration: > many "clients" syncing to one central "server". At > that time unison would create a .# temp file such that > if two unisons from different systems would be syncing > the same file at the same time they would each create > the same .# temp file and thus clobber each other. > > At that time a special unstable version existed that > would create a .# temp file and thus the > clobbering would not happen. The current stable > version does not seem to do this. Is this no longer > needed? Or do we still need to fix this. I remember this conversation from a while back, but have lost track of what happened to the fix. So I guess it needs to be fixed again. Should be a one-liner, I think... - Benjamin From hvrietsc at yahoo.com Mon Apr 4 13:43:32 2005 From: hvrietsc at yahoo.com (HvR) Date: Mon Apr 4 15:43:41 2005 Subject: [Unison-hackers] clobbering in star configurations Message-ID: <20050404194332.30494.qmail@web53707.mail.yahoo.com> > About a year or two ago I set up a star onfiguration: > many "clients" syncing to one central "server". At that time unison > would create a .# temp file such that if two unisons from different > systems would be syncing the same file at the same time they would > each create the same .# temp file and thus clobber each other. > > At that time a special unstable version existed that would create a > .# temp file and thus the clobbering would not happen. The > current stable version does not seem to do this. Is this no longer > needed? Or do we still need to fix this. I remember this conversation from a while back, but have lost track of what happened to the fix. So I guess it needs to be fixed again. Should be a one-liner, I think... - Benjamin Thanks looking forward to it. HvR __________________________________ Do you Yahoo!? Yahoo! Small Business - Try our new resources site! http://smallbusiness.yahoo.com/resources/ From bcpierce at cis.upenn.edu Mon Apr 4 22:51:57 2005 From: bcpierce at cis.upenn.edu (Benjamin Pierce) Date: Tue Apr 5 10:30:07 2005 Subject: [Unison-hackers] clobbering in star configurations In-Reply-To: Your message of Mon, 04 Apr 2005 12:43:32 -0700. <20050404194332.30494.qmail@web53707.mail.yahoo.com> Message-ID: <4640.1112647917@benjamin-pierces-computer.local> > I remember this conversation from a while back, but > have lost track of what happened to the fix. So I > guess it needs to be fixed again. Should be a > one-liner, I think... > > - Benjamin > > Thanks looking forward to it. I didn't quite say I was signing up to do it right now! I'll try to remember to get to it, but I am travelling this week and probably won't have the requisite few uninterrupted minutes. A patch would be gladly received, however... :-) - B From dbely at mail.ru Wed Apr 13 21:39:02 2005 From: dbely at mail.ru (Dmitry Bely) Date: Wed Apr 13 12:39:33 2005 Subject: [Unison-hackers] Re: [unison-users] read-only files and times=true problem under Windows; Unison crashes In-Reply-To: ("mszummer"'s message of "Tue, 29 Mar 2005 11:19:15 -0000") References: Message-ID: <64yqocm1.fsf@mail.ru> "mszummer" writes: > During four years of using Unison, I have never been able to > properly sync files including modification times (the 'times=true' > option) between two Windows XP machines both using NTFS filesystems > and synchronized clocks. I use Windows precompiled binaries (for > Windows not Cygwin) provided by links on your site. > > In version 2.9.1 and earlier, Unison would falsely list many files > as having differing modification times, claiming the difference to be > in file modification time by exactly 1 hour (even though the files > had been freshly copied from one place to another using standard > Windows copy, or also when they had previously been synced with > Unison including modification times). > There are so many files listed as differing by 1 hour that it is > difficult to pick out what files have actually changed, making the > program less usable. > > In version 2.10.2 this problem has been solved, but another and > nastier problem has appeared. The release notes describe a 'fix to > read-only under Windows' sync by Dmitry Bely. However, I experience > more problems in 2.10.2 than in 2.9.1 with regards to read-only > files, in fact I resort to 2.9.1 to get around the new problem. > > Specifically, in 2.10.2 synchronization frequently (if not always?) > fails when both files are read-only. I get the error 'Error in > setting modification times: Permission denied [utimes()] It does work > when any one of them is read-write. Well I have upgraded my Unison and now also suffer from this. I think the problem is that Unison developers have improved my old patch (though I don't know when it had happened) Let's see. I proposed the following code in props.ml: match Util.osType with `Win32 -> let oldPerms = (Unix.lstat abspath).Unix.st_perm in begin Unix.chmod abspath 0o600; Unix.utimes abspath v v; Unix.chmod abspath oldPerms end | _ -> Unix.utimes abspath v v) but now we have let iCanWrite p = try Unix.access p [Unix.R_OK]; true with Unix.Unix_error _ -> false [...] let abspath = Fspath.concatToString fspath path in if Util.osType = `Win32 && not (iCanWrite abspath) then begin (* Nb. This workaround was proposed by Dmitry Bely, to work around the fact that Unix.utimes fails on readonly files under windows. I'm [bcp] a little bit uncomfortable with it for two reasons: (1) if we crash in the middle, the permissions might be left in a bad state, and (2) I don't understand the Win32 permissions model enough to know whether it will always work -- e.g., what if the UID of the unison process is not the same as that of the file itself (under Unix, this case would fail, but we certainly don't want to make it WORLD-writable, even briefly!). *) let oldPerms = (Unix.LargeFile.lstat abspath).Unix.LargeFile.st_perm in Util.finalize (fun()-> Unix.chmod abspath 0o600; Unix.utimes abspath v v) (fun()-> Unix.chmod abspath oldPerms) end else Unix.utimes abspath v v) The comments are completely irrelevant. Windows permissions are ACL-based and not affected by Unix.chmod (i.e. C runtime chmod()) at all (see and note #define _S_IREAD 0000400 /* read permission, owner */ #define _S_IWRITE 0000200 /* write permission, owner */ definitons from SYS\STAT.H) All that chmod() affects is read-only flag which is the separate file attribute and does not relate to ACL (and thus user/group permissions) in any way. Next, "Unix.access p [Unix.R_OK];" is probably the typo: it checks if the file is readable, not writable, and the Win32 specific code in fact is never executed. "Unix.W_OK" should be used there instead. So there is no surprise that the r/o files are not synchronized correctly. I hope Unison developers are reading this and will update the sources accordingly. - Dmitry Bely From maxb at ukf.net Sun Apr 17 18:02:19 2005 From: maxb at ukf.net (Max Bowsher) Date: Sun Apr 17 12:02:29 2005 Subject: [Unison-hackers] https://cvs.cis.upenn.edu:3690 appears offline? Message-ID: <032301c54366$d5ba78e0$6508a8c0@chimaera> https://cvs.cis.upenn.edu:3690 appears offline? Intentional downtime, or malfunction? Max. From maxb at ukf.net Sun Apr 17 21:19:17 2005 From: maxb at ukf.net (Max Bowsher) Date: Sun Apr 17 15:19:33 2005 Subject: [Unison-hackers] Need a bit of ML help Message-ID: <046701c54382$59b1a130$6508a8c0@chimaera> I'm working on a win32 gtk2 gui build, and have *almost* got there! The one bizarre issue I have encountered is this, and I am hoping someone can help explain. If unison is started AND the .unison directory does not exists AND the unison.exe was compiled with the subsystem windows linker option, and thus has no attached console, THEN-AND-ONLY-THEN, the program dies immediately without any indication of what went wrong at all. This badness goes away if I comment out the following line from canonizeFspath in fspath.ml: Printf.eprintf "p = %s; bn = %s\n" p bn; flush stderr; The truly bizarre thing is that this code is executed if the .unison directory exists, too, but in that case it is not causing the program to die. Thus, my questions are: 1) Will Printf.eprintf, or flush, throw exceptions on output failure, which if uncaught, might cause the program to die without any message. 2) Is canonizeFspath called in different circumstances depending on whether the .unison directory exists or not, and might one be catching exceptions that the other does not? Max. From bcpierce at cis.upenn.edu Sun Apr 17 20:35:28 2005 From: bcpierce at cis.upenn.edu (Benjamin Pierce) Date: Sun Apr 17 19:35:37 2005 Subject: [Unison-hackers] https://cvs.cis.upenn.edu:3690 appears offline? In-Reply-To: Your message of Sun, 17 Apr 2005 17:02:19 +0100. <032301c54366$d5ba78e0$6508a8c0@chimaera> Message-ID: <18316.1113780928@dsl-207-245-69-66.cust.oldcity.dca.net> Geoff tells me he's restarted it now. - B From maxb at ukf.net Mon Apr 18 11:40:45 2005 From: maxb at ukf.net (Max Bowsher) Date: Mon Apr 18 05:40:57 2005 Subject: [Unison-hackers] https://cvs.cis.upenn.edu:3690 appears offline, again! Message-ID: <000d01c543fa$b275b270$6508a8c0@chimaera> Thanks for getting the server up, but now it appears to be down again! Max. From maxb at ukf.net Mon Apr 18 13:59:55 2005 From: maxb at ukf.net (Max Bowsher) Date: Mon Apr 18 08:00:11 2005 Subject: [Unison-hackers] Re: https://cvs.cis.upenn.edu:3690 appears offline, again! References: <000d01c543fa$b275b270$6508a8c0@chimaera> Message-ID: <00e501c5440e$22df5620$6508a8c0@chimaera> Geoffrey A. Washburn wrote: > On Mon, 18 Apr 2005, Max Bowsher wrote: > >> Thanks for getting the server up, but now it appears to be down again! > > It seems fine on my end (machine is up, daemon running). Have you > tried tracerouting to see if it some kind of network fault? It's fine again for me too now. I don't know why I was getting 'connection refused' again for a while. Max. From maxb at ukf.net Mon Apr 18 15:04:46 2005 From: maxb at ukf.net (Max Bowsher) Date: Mon Apr 18 09:04:53 2005 Subject: [Unison-hackers] Unison Windows Binaries Message-ID: <01b201c54417$326eb5a0$6508a8c0@chimaera> Windows binaries are now available in the contributed binaries repository: https://cvs.cis.upenn.edu:3690/svnroot/unison-contributed-binaries/windows/ Versions: Current Beta: 2.12.0 (text + gtk2) Current Stable: 2.10.2 (text + gtk2) Previous Beta: 2.9.20 (text) Previous Stable: 2.9.1 (text) Max. From bcpierce at cis.upenn.edu Mon Apr 18 10:18:12 2005 From: bcpierce at cis.upenn.edu (Benjamin Pierce) Date: Mon Apr 18 09:18:30 2005 Subject: [Unison-hackers] Unison Windows Binaries In-Reply-To: Your message of Mon, 18 Apr 2005 14:04:46 +0100. <01b201c54417$326eb5a0$6508a8c0@chimaera> Message-ID: <18840.1113830292@dsl-207-245-69-66.cust.oldcity.dca.net> Thanks, Max -- I was able to grab them with no problem. I have one question, though: what's the best way to make sure that people downloading these files look at the README and notice the dependencies that they need to satisfy for the GTK+ versions? The simplest idea is to just keep all these files together in a directory on the Unison download site and make a separate button on the download page that takes you to that directory. Does that seem reasonable? - Benjamin From trevor at research.att.com Mon Apr 18 11:34:27 2005 From: trevor at research.att.com (Trevor Jim) Date: Mon Apr 18 10:31:19 2005 Subject: [Unison-hackers] Need a bit of ML help In-Reply-To: <046701c54382$59b1a130$6508a8c0@chimaera> References: <046701c54382$59b1a130$6508a8c0@chimaera> Message-ID: On Apr 17, 2005, at 3:19 PM, Max Bowsher wrote: > This badness goes away if I comment out the following line from > canonizeFspath in fspath.ml: > Printf.eprintf "p = %s; bn = %s\n" p bn; flush stderr; I'm not sure what the problem is. But, I would be happy if this line was simply deleted. I think there are one or two places like this where essentially someone put in some debugging code but forgot to take it out on a checkin. I don't think this message is helpful for the end user. Anyone remember why this is there? Getting back to the cause of the crash: I suspect that the linker option is causing a problem for ocaml. Perhaps with that linker option some ocaml initialization code does not get run properly. Jerome might know more. -Trevor From maxb at ukf.net Mon Apr 18 20:50:50 2005 From: maxb at ukf.net (Max Bowsher) Date: Mon Apr 18 14:51:02 2005 Subject: [Unison-hackers] Unison Windows Binaries References: <18840.1113830292@dsl-207-245-69-66.cust.oldcity.dca.net> Message-ID: <03f201c54447$8aa1bf30$6508a8c0@chimaera> Benjamin Pierce wrote: > Thanks, Max -- I was able to grab them with no problem. I have one > question, though: what's the best way to make sure that people downloading > these files look at the README and notice the dependencies that they need > to > satisfy for the GTK+ versions? > > The simplest idea is to just keep all these files together in a directory > on > the Unison download site and make a separate button on the download page > that takes you to that directory. Does that seem reasonable? Yes, definitely. You could put "Make sure you read the README file" in bold red text, next to the button, just to emphasize the issue. Max. From bcpierce at cis.upenn.edu Wed Apr 20 10:39:54 2005 From: bcpierce at cis.upenn.edu (Benjamin Pierce) Date: Wed Apr 20 09:40:12 2005 Subject: [Unison-hackers] Unison Windows Binaries In-Reply-To: Your message of Mon, 18 Apr 2005 19:50:50 +0100. <03f201c54447$8aa1bf30$6508a8c0@chimaera> Message-ID: <21268.1114004394@dsl-207-245-69-66.cust.oldcity.dca.net> > > The simplest idea is to just keep all these files together in a directory > > on > > the Unison download site and make a separate button on the download page > > that takes you to that directory. Does that seem reasonable? > > Yes, definitely. > > You could put "Make sure you read the README file" in bold red text, next to > the button, just to emphasize the issue. OK, done. I'll make an announcement to the users lists shortly. - Benjamin From maxb at ukf.net Wed Apr 20 15:56:46 2005 From: maxb at ukf.net (Max Bowsher) Date: Wed Apr 20 09:57:04 2005 Subject: [Unison-hackers] Unison Windows Binaries References: <21268.1114004394@dsl-207-245-69-66.cust.oldcity.dca.net> Message-ID: <03d301c545b0$cab705c0$6508a8c0@chimaera> Benjamin Pierce wrote: >>> The simplest idea is to just keep all these files together in a >>> directory >>> on >>> the Unison download site and make a separate button on the download page >>> that takes you to that directory. Does that seem reasonable? >> >> Yes, definitely. >> >> You could put "Make sure you read the README file" in bold red text, next >> to the button, just to emphasize the issue. > > OK, done. I'll make an announcement to the users lists shortly. At the moment, the "View contributed ports" button is broken - it links to the dev-snapshot instead! Max. From bcpierce at cis.upenn.edu Wed Apr 20 14:46:12 2005 From: bcpierce at cis.upenn.edu (Benjamin Pierce) Date: Wed Apr 20 13:46:08 2005 Subject: [Unison-hackers] Unison Windows Binaries In-Reply-To: Your message of Wed, 20 Apr 2005 14:56:46 +0100. <03d301c545b0$cab705c0$6508a8c0@chimaera> Message-ID: <667.1114019172@dhcp38-120.seas.upenn.edu> > At the moment, the "View contributed ports" button is broken - it links to > the dev-snapshot instead! Reload the web page -- the name of the button has changed! - B From maxb at ukf.net Thu Apr 21 00:17:56 2005 From: maxb at ukf.net (Max Bowsher) Date: Wed Apr 20 18:18:20 2005 Subject: [Unison-hackers] Unison Windows Binaries References: <667.1114019172@dhcp38-120.seas.upenn.edu> Message-ID: <069b01c545f6$ce2bb750$6508a8c0@chimaera> Benjamin Pierce wrote: >> At the moment, the "View contributed ports" button is broken - it links >> to >> the dev-snapshot instead! > > Reload the web page -- the name of the button has changed! Ah! Could you remove my list item from contrib.html now, please, since it is now out of date. Max. From bcpierce at cis.upenn.edu Wed Apr 20 20:03:27 2005 From: bcpierce at cis.upenn.edu (Benjamin Pierce) Date: Wed Apr 20 19:03:37 2005 Subject: [Unison-hackers] Unison Windows Binaries In-Reply-To: Your message of Wed, 20 Apr 2005 23:17:56 +0100. <069b01c545f6$ce2bb750$6508a8c0@chimaera> Message-ID: <21479.1114038207@dsl-207-245-69-66.cust.oldcity.dca.net> > Could you remove my list item from contrib.html now, please, since it is now > out of date. Done. - B From subversion at central.cis.upenn.edu Mon Apr 25 14:19:59 2005 From: subversion at central.cis.upenn.edu (subversion@central.cis.upenn.edu) Date: Mon Apr 25 13:20:05 2005 Subject: [Unison-hackers] [unison-svn] rev 27 - in trunk: . doc src src/ubase Message-ID: <200504251719.j3PHJxtt002145@tuber.cis.upenn.edu> Author: bcpierce Date: 2005-04-25 13:19:57 -0400 (Mon, 25 Apr 2005) New Revision: 27 Modified: trunk/.depend trunk/doc/contactsbody.tex trunk/src/Makefile trunk/src/RECENTNEWS trunk/src/TODO.txt trunk/src/common.ml trunk/src/files.ml trunk/src/globals.ml trunk/src/mkProjectInfo.ml trunk/src/os.ml trunk/src/osx.ml trunk/src/path.ml trunk/src/pred.ml trunk/src/pred.mli trunk/src/props.ml trunk/src/recon.ml trunk/src/remote.ml trunk/src/terminal.ml trunk/src/ubase/depend trunk/src/ubase/prefs.ml trunk/src/ubase/prefs.mli trunk/src/ubase/util.ml trunk/src/uicommon.ml trunk/src/uicommon.mli trunk/src/uigtk.ml trunk/src/uigtk2.ml trunk/src/uitext.ml trunk/src/update.ml Log: * Changed time display to include seconds (so that people on FAT filesystems will not be confused when Unison tries to update a file time to an odd number of seconds and the filesystem truncates it to an even number!) * Changed some occurrences of List.foo to Safelist.foo (just for paranoia) * Added Prefs.name function (actually needed by Harmony) and changed type of Prefs.alias to take a Prefs.t instead of a string as first parameter (for more safety) * WARNING: everything compiles, but the new Prefs module has not been tested yet. (I'll do that soon.) From maxb at ukf.net Thu Apr 28 13:32:11 2005 From: maxb at ukf.net (Max Bowsher) Date: Thu Apr 28 07:32:16 2005 Subject: [Unison-hackers] Windows read-only files issue reported by Dmitry Bely Message-ID: <02e601c54be5$eb6bed40$e6ec6f83@chimaera> I was recently contacted privately asking about any progress regarding the following: http://groups.yahoo.com/group/unison-users/message/3667 Given that the original problem report was on unison-users, not unison-hackers, I think it quite likely that it was overlooked. In particular, it looks very suspicious that the function iCanWrite is testing for _read_ permission! Max.